[Qemu-devel] [PULL 0/4] Net patches
[PULL 0/5] Net patches
1
The following changes since commit a04d91c701251a9b32b7364ddb48029ba024cb75:
1
The following changes since commit 23895cbd82be95428e90168b12e925d0d3ca2f06:
2
2
3
Merge remote-tracking branch 'remotes/alistair/tags/pull-device-tree-20190327' into staging (2019-03-28 12:39:43 +0000)
3
Merge remote-tracking branch 'remotes/awilliam/tags/vfio-update-20201123.0' into staging (2020-11-23 18:51:13 +0000)
4
4
5
are available in the git repository at:
5
are available in the git repository at:
6
6
7
https://github.com/jasowang/qemu.git tags/net-pull-request
7
https://github.com/jasowang/qemu.git tags/net-pull-request
8
8
9
for you to fetch changes up to ab79237a15e8f8c23310291b672d83374cf17935:
9
for you to fetch changes up to 9925990d01a92564af55f6f69d0f5f59b47609b1:
10
10
11
net: tap: use qemu_set_nonblock (2019-03-29 15:22:18 +0800)
11
net: Use correct default-path macro for downscript (2020-11-24 10:40:17 +0800)
12
12
13
----------------------------------------------------------------
13
----------------------------------------------------------------
14
14
15
----------------------------------------------------------------
15
----------------------------------------------------------------
16
Li Qiang (1):
16
Keqian Zhu (1):
17
net: tap: use qemu_set_nonblock
17
net: Use correct default-path macro for downscript
18
18
19
Marc-André Lureau (1):
19
Paolo Bonzini (1):
20
net/socket: learn to talk with a unix dgram socket
20
net: do not exit on "netdev_add help" monitor command
21
21
22
Zhang Chen (1):
22
Prasad J Pandit (1):
23
MAINTAINERS: Update the latest email address
23
hw/net/e1000e: advance desc_offset in case of null descriptor
24
24
25
yuchenlin (1):
25
Yuri Benditovich (1):
26
e1000: Delay flush queue when receive RCTL
26
net: purge queued rx packets on queue deletion
27
27
28
MAINTAINERS | 2 +-
28
yuanjungong (1):
29
hw/net/e1000.c | 24 ++++++++++++++++++++++--
29
tap: fix a memory leak
30
net/socket.c | 25 +++++++++++++++++++++----
30
31
net/tap.c | 10 +++++-----
31
hw/net/e1000e_core.c | 8 +++---
32
4 files changed, 49 insertions(+), 12 deletions(-)
32
include/net/net.h | 1 +
33
monitor/hmp-cmds.c | 6 ++++
34
net/net.c | 80 +++++++++++++++++++++++++++-------------------------
35
net/tap.c | 5 +++-
36
5 files changed, 57 insertions(+), 43 deletions(-)
33
37
34
38
diff view generated by jsdifflib
New patch
[PULL 1/5] hw/net/e1000e: advance desc_offset in case of null descriptor
1
From: Prasad J Pandit <pjp@fedoraproject.org>
1
2
3
While receiving packets via e1000e_write_packet_to_guest() routine,
4
'desc_offset' is advanced only when RX descriptor is processed. And
5
RX descriptor is not processed if it has NULL buffer address.
6
This may lead to an infinite loop condition. Increament 'desc_offset'
7
to process next descriptor in the ring to avoid infinite loop.
8
9
Reported-by: Cheol-woo Myung <330cjfdn@gmail.com>
10
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
11
Signed-off-by: Jason Wang <jasowang@redhat.com>
12
---
13
hw/net/e1000e_core.c | 8 ++++----
14
1 file changed, 4 insertions(+), 4 deletions(-)
15
16
diff --git a/hw/net/e1000e_core.c b/hw/net/e1000e_core.c
17
index XXXXXXX..XXXXXXX 100644
18
--- a/hw/net/e1000e_core.c
19
+++ b/hw/net/e1000e_core.c
20
@@ -XXX,XX +XXX,XX @@ e1000e_write_packet_to_guest(E1000ECore *core, struct NetRxPkt *pkt,
21
(const char *) &fcs_pad, e1000x_fcs_len(core->mac));
22
}
23
}
24
- desc_offset += desc_size;
25
- if (desc_offset >= total_size) {
26
- is_last = true;
27
- }
28
} else { /* as per intel docs; skip descriptors with null buf addr */
29
trace_e1000e_rx_null_descriptor();
30
}
31
+ desc_offset += desc_size;
32
+ if (desc_offset >= total_size) {
33
+ is_last = true;
34
+ }
35
36
e1000e_write_rx_descr(core, desc, is_last ? core->rx_pkt : NULL,
37
rss_info, do_ps ? ps_hdr_len : 0, &bastate.written);
38
--
39
2.7.4
40
41
diff view generated by jsdifflib
[Qemu-devel] [PULL 2/4] e1000: Delay flush queue when receive RCTL
[PULL 2/5] net: do not exit on "netdev_add help" monitor command
1
From: yuchenlin <yuchenlin@synology.com>
1
From: Paolo Bonzini <pbonzini@redhat.com>
2
2
3
Due to too early RCT0 interrput, win10x32 may hang on booting.
3
"netdev_add help" is causing QEMU to exit because the code that
4
This problem can be reproduced by doing power cycle on win10x32 guest.
4
invokes show_netdevs is shared between CLI and HMP processing.
5
In our environment, we have 10 win10x32 and stress power cycle.
5
Move the check to the callers so that exit(0) remains only
6
The problem will happen about 20 rounds.
6
in the CLI flow.
7
7
8
Below shows some log with comment:
8
"netdev_add help" is not fixed by this patch; that is left for
9
later work.
9
10
10
The normal case:
11
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
11
12
22831@1551928392.984687:e1000x_rx_disabled Received packet dropped
13
because receive is disabled RCTL = 0
14
22831@1551928392.985655:e1000x_rx_disabled Received packet dropped
15
because receive is disabled RCTL = 0
16
22831@1551928392.985801:e1000x_rx_disabled Received packet dropped
17
because receive is disabled RCTL = 0
18
e1000: set_ics 0, ICR 0, IMR 0
19
e1000: set_ics 0, ICR 0, IMR 0
20
e1000: set_ics 0, ICR 0, IMR 0
21
e1000: RCTL: 0, mac_reg[RCTL] = 0x0
22
22831@1551928393.056710:e1000x_rx_disabled Received packet dropped
23
because receive is disabled RCTL = 0
24
e1000: set_ics 0, ICR 0, IMR 0
25
e1000: ICR read: 0
26
e1000: set_ics 0, ICR 0, IMR 0
27
e1000: set_ics 0, ICR 0, IMR 0
28
e1000: RCTL: 0, mac_reg[RCTL] = 0x0
29
22831@1551928393.077548:e1000x_rx_disabled Received packet dropped
30
because receive is disabled RCTL = 0
31
e1000: set_ics 0, ICR 0, IMR 0
32
e1000: ICR read: 0
33
e1000: set_ics 2, ICR 0, IMR 0
34
e1000: set_ics 2, ICR 2, IMR 0
35
e1000: RCTL: 0, mac_reg[RCTL] = 0x0
36
22831@1551928393.102974:e1000x_rx_disabled Received packet dropped
37
because receive is disabled RCTL = 0
38
22831@1551928393.103267:e1000x_rx_disabled Received packet dropped
39
because receive is disabled RCTL = 0
40
e1000: RCTL: 255, mac_reg[RCTL] = 0x40002 <- win10x32 says it can handle
41
RX now
42
e1000: set_ics 0, ICR 2, IMR 9d <- unmask interrupt
43
e1000: RCTL: 255, mac_reg[RCTL] = 0x48002
44
e1000: set_ics 80, ICR 2, IMR 9d <- interrupt and work!
45
...
46
47
The bad case:
48
49
27744@1551930483.117766:e1000x_rx_disabled Received packet dropped
50
because receive is disabled RCTL = 0
51
27744@1551930483.118398:e1000x_rx_disabled Received packet dropped
52
because receive is disabled RCTL = 0
53
e1000: set_ics 0, ICR 0, IMR 0
54
e1000: set_ics 0, ICR 0, IMR 0
55
e1000: set_ics 0, ICR 0, IMR 0
56
e1000: RCTL: 0, mac_reg[RCTL] = 0x0
57
27744@1551930483.198063:e1000x_rx_disabled Received packet dropped
58
because receive is disabled RCTL = 0
59
e1000: set_ics 0, ICR 0, IMR 0
60
e1000: ICR read: 0
61
e1000: set_ics 0, ICR 0, IMR 0
62
e1000: set_ics 0, ICR 0, IMR 0
63
e1000: RCTL: 0, mac_reg[RCTL] = 0x0
64
27744@1551930483.218675:e1000x_rx_disabled Received packet dropped
65
because receive is disabled RCTL = 0
66
e1000: set_ics 0, ICR 0, IMR 0
67
e1000: ICR read: 0
68
e1000: set_ics 2, ICR 0, IMR 0
69
e1000: set_ics 2, ICR 2, IMR 0
70
e1000: RCTL: 0, mac_reg[RCTL] = 0x0
71
27744@1551930483.241768:e1000x_rx_disabled Received packet dropped
72
because receive is disabled RCTL = 0
73
27744@1551930483.241979:e1000x_rx_disabled Received packet dropped
74
because receive is disabled RCTL = 0
75
e1000: RCTL: 255, mac_reg[RCTL] = 0x40002 <- win10x32 says it can handle
76
RX now
77
e1000: set_ics 80, ICR 2, IMR 0 <- flush queue (caused by setting RCTL)
78
e1000: set_ics 0, ICR 82, IMR 9d <- unmask interrupt and because 0x82&0x9d
79
!= 0 generate interrupt, hang on here...
80
81
To workaround this problem, simply delay flush queue. Also stop receiving
82
when timer is going to run.
83
84
Tested on CentOS, Win7SP1x64 and Win10x32.
85
86
Signed-off-by: yuchenlin <yuchenlin@synology.com>
87
Reviewed-by: Dmitry Fleytman <dmitry.fleytman@gmail.com>
88
Signed-off-by: Jason Wang <jasowang@redhat.com>
12
Signed-off-by: Jason Wang <jasowang@redhat.com>
89
---
13
---
90
hw/net/e1000.c | 24 ++++++++++++++++++++++--
14
include/net/net.h | 1 +
91
1 file changed, 22 insertions(+), 2 deletions(-)
15
monitor/hmp-cmds.c | 6 +++++
16
net/net.c | 68 +++++++++++++++++++++++++++---------------------------
17
3 files changed, 41 insertions(+), 34 deletions(-)
92
18
93
diff --git a/hw/net/e1000.c b/hw/net/e1000.c
19
diff --git a/include/net/net.h b/include/net/net.h
94
index XXXXXXX..XXXXXXX 100644
20
index XXXXXXX..XXXXXXX 100644
95
--- a/hw/net/e1000.c
21
--- a/include/net/net.h
96
+++ b/hw/net/e1000.c
22
+++ b/include/net/net.h
97
@@ -XXX,XX +XXX,XX @@ typedef struct E1000State_st {
23
@@ -XXX,XX +XXX,XX @@ extern const char *host_net_devices[];
98
bool mit_irq_level; /* Tracks interrupt pin level. */
24
99
uint32_t mit_ide; /* Tracks E1000_TXD_CMD_IDE bit. */
25
/* from net.c */
100
26
int net_client_parse(QemuOptsList *opts_list, const char *str);
101
+ QEMUTimer *flush_queue_timer;
27
+void show_netdevs(void);
28
int net_init_clients(Error **errp);
29
void net_check_clients(void);
30
void net_cleanup(void);
31
diff --git a/monitor/hmp-cmds.c b/monitor/hmp-cmds.c
32
index XXXXXXX..XXXXXXX 100644
33
--- a/monitor/hmp-cmds.c
34
+++ b/monitor/hmp-cmds.c
35
@@ -XXX,XX +XXX,XX @@
36
#include "qemu/option.h"
37
#include "qemu/timer.h"
38
#include "qemu/sockets.h"
39
+#include "qemu/help_option.h"
40
#include "monitor/monitor-internal.h"
41
#include "qapi/error.h"
42
#include "qapi/clone-visitor.h"
43
@@ -XXX,XX +XXX,XX @@ void hmp_netdev_add(Monitor *mon, const QDict *qdict)
44
{
45
Error *err = NULL;
46
QemuOpts *opts;
47
+ const char *type = qdict_get_try_str(qdict, "type");
48
49
+ if (type && is_help_option(type)) {
50
+ show_netdevs();
51
+ return;
52
+ }
53
opts = qemu_opts_from_qdict(qemu_find_opts("netdev"), qdict, &err);
54
if (err) {
55
goto out;
56
diff --git a/net/net.c b/net/net.c
57
index XXXXXXX..XXXXXXX 100644
58
--- a/net/net.c
59
+++ b/net/net.c
60
@@ -XXX,XX +XXX,XX @@
61
#include "qemu/config-file.h"
62
#include "qemu/ctype.h"
63
#include "qemu/iov.h"
64
+#include "qemu/qemu-print.h"
65
#include "qemu/main-loop.h"
66
#include "qemu/option.h"
67
#include "qapi/error.h"
68
@@ -XXX,XX +XXX,XX @@ static int net_client_init1(const Netdev *netdev, bool is_netdev, Error **errp)
69
return 0;
70
}
71
72
-static void show_netdevs(void)
73
+void show_netdevs(void)
74
{
75
int idx;
76
const char *available_netdevs[] = {
77
@@ -XXX,XX +XXX,XX @@ static void show_netdevs(void)
78
#endif
79
};
80
81
- printf("Available netdev backend types:\n");
82
+ qemu_printf("Available netdev backend types:\n");
83
for (idx = 0; idx < ARRAY_SIZE(available_netdevs); idx++) {
84
- puts(available_netdevs[idx]);
85
+ qemu_printf("%s\n", available_netdevs[idx]);
86
}
87
}
88
89
@@ -XXX,XX +XXX,XX @@ static int net_client_init(QemuOpts *opts, bool is_netdev, Error **errp)
90
int ret = -1;
91
Visitor *v = opts_visitor_new(opts);
92
93
- const char *type = qemu_opt_get(opts, "type");
94
-
95
- if (is_netdev && type && is_help_option(type)) {
96
- show_netdevs();
97
- exit(0);
98
- } else {
99
- /* Parse convenience option format ip6-net=fec0::0[/64] */
100
- const char *ip6_net = qemu_opt_get(opts, "ipv6-net");
101
+ /* Parse convenience option format ip6-net=fec0::0[/64] */
102
+ const char *ip6_net = qemu_opt_get(opts, "ipv6-net");
103
104
- if (ip6_net) {
105
- char *prefix_addr;
106
- unsigned long prefix_len = 64; /* Default 64bit prefix length. */
107
+ if (ip6_net) {
108
+ char *prefix_addr;
109
+ unsigned long prefix_len = 64; /* Default 64bit prefix length. */
110
111
- substrings = g_strsplit(ip6_net, "/", 2);
112
- if (!substrings || !substrings[0]) {
113
- error_setg(errp, QERR_INVALID_PARAMETER_VALUE, "ipv6-net",
114
- "a valid IPv6 prefix");
115
- goto out;
116
- }
117
+ substrings = g_strsplit(ip6_net, "/", 2);
118
+ if (!substrings || !substrings[0]) {
119
+ error_setg(errp, QERR_INVALID_PARAMETER_VALUE, "ipv6-net",
120
+ "a valid IPv6 prefix");
121
+ goto out;
122
+ }
123
124
- prefix_addr = substrings[0];
125
+ prefix_addr = substrings[0];
126
127
- /* Handle user-specified prefix length. */
128
- if (substrings[1] &&
129
- qemu_strtoul(substrings[1], NULL, 10, &prefix_len))
130
- {
131
- error_setg(errp, QERR_INVALID_PARAMETER_VALUE,
132
- "ipv6-prefixlen", "a number");
133
- goto out;
134
- }
135
-
136
- qemu_opt_set(opts, "ipv6-prefix", prefix_addr, &error_abort);
137
- qemu_opt_set_number(opts, "ipv6-prefixlen", prefix_len,
138
- &error_abort);
139
- qemu_opt_unset(opts, "ipv6-net");
140
+ /* Handle user-specified prefix length. */
141
+ if (substrings[1] &&
142
+ qemu_strtoul(substrings[1], NULL, 10, &prefix_len))
143
+ {
144
+ error_setg(errp, QERR_INVALID_PARAMETER_VALUE,
145
+ "ipv6-prefixlen", "a number");
146
+ goto out;
147
}
102
+
148
+
103
/* Compatibility flags for migration to/from qemu 1.3.0 and older */
149
+ qemu_opt_set(opts, "ipv6-prefix", prefix_addr, &error_abort);
104
#define E1000_FLAG_AUTONEG_BIT 0
150
+ qemu_opt_set_number(opts, "ipv6-prefixlen", prefix_len,
105
#define E1000_FLAG_MIT_BIT 1
151
+ &error_abort);
106
@@ -XXX,XX +XXX,XX @@ static void e1000_reset(void *opaque)
152
+ qemu_opt_unset(opts, "ipv6-net");
107
153
}
108
timer_del(d->autoneg_timer);
154
109
timer_del(d->mit_timer);
155
/* Create an ID for -net if the user did not specify one */
110
+ timer_del(d->flush_queue_timer);
156
@@ -XXX,XX +XXX,XX @@ static int net_init_client(void *dummy, QemuOpts *opts, Error **errp)
111
d->mit_timer_on = 0;
157
112
d->mit_irq_level = 0;
158
static int net_init_netdev(void *dummy, QemuOpts *opts, Error **errp)
113
d->mit_ide = 0;
159
{
114
@@ -XXX,XX +XXX,XX @@ set_ctrl(E1000State *s, int index, uint32_t val)
160
+ const char *type = qemu_opt_get(opts, "type");
161
+
162
+ if (type && is_help_option(type)) {
163
+ show_netdevs();
164
+ exit(0);
165
+ }
166
return net_client_init(opts, true, errp);
115
}
167
}
116
168
117
static void
118
+e1000_flush_queue_timer(void *opaque)
119
+{
120
+ E1000State *s = opaque;
121
+
122
+ qemu_flush_queued_packets(qemu_get_queue(s->nic));
123
+}
124
+
125
+static void
126
set_rx_control(E1000State *s, int index, uint32_t val)
127
{
128
s->mac_reg[RCTL] = val;
129
@@ -XXX,XX +XXX,XX @@ set_rx_control(E1000State *s, int index, uint32_t val)
130
s->rxbuf_min_shift = ((val / E1000_RCTL_RDMTS_QUAT) & 3) + 1;
131
DBGOUT(RX, "RCTL: %d, mac_reg[RCTL] = 0x%x\n", s->mac_reg[RDT],
132
s->mac_reg[RCTL]);
133
- qemu_flush_queued_packets(qemu_get_queue(s->nic));
134
+ timer_mod(s->flush_queue_timer,
135
+ qemu_clock_get_ms(QEMU_CLOCK_VIRTUAL) + 1000);
136
}
137
138
static void
139
@@ -XXX,XX +XXX,XX @@ e1000_can_receive(NetClientState *nc)
140
E1000State *s = qemu_get_nic_opaque(nc);
141
142
return e1000x_rx_ready(&s->parent_obj, s->mac_reg) &&
143
- e1000_has_rxbufs(s, 1);
144
+ e1000_has_rxbufs(s, 1) && !timer_pending(s->flush_queue_timer);
145
}
146
147
static uint64_t rx_desc_base(E1000State *s)
148
@@ -XXX,XX +XXX,XX @@ e1000_receive_iov(NetClientState *nc, const struct iovec *iov, int iovcnt)
149
return -1;
150
}
151
152
+ if (timer_pending(s->flush_queue_timer)) {
153
+ return 0;
154
+ }
155
+
156
/* Pad to minimum Ethernet frame length */
157
if (size < sizeof(min_buf)) {
158
iov_to_buf(iov, iovcnt, 0, min_buf, size);
159
@@ -XXX,XX +XXX,XX @@ pci_e1000_uninit(PCIDevice *dev)
160
timer_free(d->autoneg_timer);
161
timer_del(d->mit_timer);
162
timer_free(d->mit_timer);
163
+ timer_del(d->flush_queue_timer);
164
+ timer_free(d->flush_queue_timer);
165
qemu_del_nic(d->nic);
166
}
167
168
@@ -XXX,XX +XXX,XX @@ static void pci_e1000_realize(PCIDevice *pci_dev, Error **errp)
169
170
d->autoneg_timer = timer_new_ms(QEMU_CLOCK_VIRTUAL, e1000_autoneg_timer, d);
171
d->mit_timer = timer_new_ns(QEMU_CLOCK_VIRTUAL, e1000_mit_timer, d);
172
+ d->flush_queue_timer = timer_new_ms(QEMU_CLOCK_VIRTUAL,
173
+ e1000_flush_queue_timer, d);
174
}
175
176
static void qdev_e1000_reset(DeviceState *dev)
177
--
169
--
178
2.5.0
170
2.7.4
179
171
180
172
diff view generated by jsdifflib
[Qemu-devel] [PULL 1/4] net/socket: learn to talk with a unix dgram socket
[PULL 3/5] net: purge queued rx packets on queue deletion
1
From: Marc-André Lureau <marcandre.lureau@redhat.com>
1
From: Yuri Benditovich <yuri.benditovich@daynix.com>
2
2
3
-net socket has a fd argument, and may be passed pre-opened sockets.
3
https://bugzilla.redhat.com/show_bug.cgi?id=1829272
4
When deleting queue pair, purge pending RX packets if any.
5
Example of problematic flow:
6
1. Bring up q35 VM with tap (vhost off) and virtio-net or e1000e
7
2. Run ping flood to the VM NIC ( 1 ms interval)
8
3. Hot unplug the NIC device (device_del)
9
During unplug process one or more packets come, the NIC
10
can't receive, tap disables read_poll
11
4. Hot plug the device (device_add) with the same netdev
12
The tap stays with read_poll disabled and does not receive
13
any packets anymore (tap_send never triggered)
4
14
5
TCP sockets use framing.
15
Signed-off-by: Yuri Benditovich <yuri.benditovich@daynix.com>
6
UDP sockets have datagram boundaries.
7
8
When given a unix dgram socket, it will be able to read from it, but
9
will attempt to send on the dgram_dst, which is unset. The other end
10
will not receive the data.
11
12
Let's teach -net socket to recognize a UNIX DGRAM socket, and use the
13
regular send() command (without dgram_dst).
14
15
This makes running slirp out-of-process possible that
16
way (python pseudo-code):
17
18
a, b = socket.socketpair(socket.AF_UNIX, socket.SOCK_DGRAM)
19
20
subprocess.Popen('qemu -net socket,fd=%d -net user' % a.fileno(), shell=True)
21
subprocess.Popen('qemu ... -net nic -net socket,fd=%d' % b.fileno(), shell=True)
22
23
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
24
Signed-off-by: Jason Wang <jasowang@redhat.com>
16
Signed-off-by: Jason Wang <jasowang@redhat.com>
25
---
17
---
26
net/socket.c | 25 +++++++++++++++++++++----
18
net/net.c | 12 ++++++++----
27
1 file changed, 21 insertions(+), 4 deletions(-)
19
1 file changed, 8 insertions(+), 4 deletions(-)
28
20
29
diff --git a/net/socket.c b/net/socket.c
21
diff --git a/net/net.c b/net/net.c
30
index XXXXXXX..XXXXXXX 100644
22
index XXXXXXX..XXXXXXX 100644
31
--- a/net/socket.c
23
--- a/net/net.c
32
+++ b/net/socket.c
24
+++ b/net/net.c
33
@@ -XXX,XX +XXX,XX @@ static ssize_t net_socket_receive_dgram(NetClientState *nc, const uint8_t *buf,
25
@@ -XXX,XX +XXX,XX @@ void qemu_del_nic(NICState *nic)
34
ssize_t ret;
26
35
27
qemu_macaddr_set_free(&nic->conf->macaddr);
36
do {
28
37
- ret = qemu_sendto(s->fd, buf, size, 0,
29
- /* If this is a peer NIC and peer has already been deleted, free it now. */
38
- (struct sockaddr *)&s->dgram_dst,
30
- if (nic->peer_deleted) {
39
- sizeof(s->dgram_dst));
31
- for (i = 0; i < queues; i++) {
40
+ if (s->dgram_dst.sin_family != AF_UNIX) {
32
- qemu_free_net_client(qemu_get_subqueue(nic, i)->peer);
41
+ ret = qemu_sendto(s->fd, buf, size, 0,
33
+ for (i = 0; i < queues; i++) {
42
+ (struct sockaddr *)&s->dgram_dst,
34
+ NetClientState *nc = qemu_get_subqueue(nic, i);
43
+ sizeof(s->dgram_dst));
35
+ /* If this is a peer NIC and peer has already been deleted, free it now. */
44
+ } else {
36
+ if (nic->peer_deleted) {
45
+ ret = send(s->fd, buf, size, 0);
37
+ qemu_free_net_client(nc->peer);
46
+ }
38
+ } else if (nc->peer) {
47
} while (ret == -1 && errno == EINTR);
39
+ /* if there are RX packets pending, complete them */
48
40
+ qemu_purge_queued_packets(nc->peer);
49
if (ret == -1 && errno == EAGAIN) {
41
}
50
@@ -XXX,XX +XXX,XX @@ static NetSocketState *net_socket_fd_init_dgram(NetClientState *peer,
51
int newfd;
52
NetClientState *nc;
53
NetSocketState *s;
54
+ SocketAddress *sa;
55
+ SocketAddressType sa_type;
56
+
57
+ sa = socket_local_address(fd, errp);
58
+ if (!sa) {
59
+ return NULL;
60
+ }
61
+ sa_type = sa->type;
62
+ qapi_free_SocketAddress(sa);
63
64
/* fd passed: multicast: "learn" dgram_dst address from bound address and save it
65
* Because this may be "shared" socket from a "master" process, datagrams would be recv()
66
@@ -XXX,XX +XXX,XX @@ static NetSocketState *net_socket_fd_init_dgram(NetClientState *peer,
67
"socket: fd=%d (cloned mcast=%s:%d)",
68
fd, inet_ntoa(saddr.sin_addr), ntohs(saddr.sin_port));
69
} else {
70
+ if (sa_type == SOCKET_ADDRESS_TYPE_UNIX) {
71
+ s->dgram_dst.sin_family = AF_UNIX;
72
+ }
73
+
74
snprintf(nc->info_str, sizeof(nc->info_str),
75
- "socket: fd=%d", fd);
76
+ "socket: fd=%d %s", fd, SocketAddressType_str(sa_type));
77
}
42
}
78
43
79
return s;
80
--
44
--
81
2.5.0
45
2.7.4
82
46
83
47
diff view generated by jsdifflib
[Qemu-devel] [PULL 4/4] net: tap: use qemu_set_nonblock
[PULL 4/5] tap: fix a memory leak
1
From: Li Qiang <liq3ea@gmail.com>
1
From: yuanjungong <ruc_gongyuanjun@163.com>
2
2
3
The fcntl will change the flags directly, use qemu_set_nonblock()
3
Close fd before returning.
4
instead.
5
4
6
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
5
Buglink: https://bugs.launchpad.net/qemu/+bug/1904486
7
Acked-by: Michael S. Tsirkin <mst@redhat.com>
6
8
Signed-off-by: Li Qiang <liq3ea@gmail.com>
7
Signed-off-by: yuanjungong <ruc_gongyuanjun@163.com>
8
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
9
Signed-off-by: Jason Wang <jasowang@redhat.com>
9
Signed-off-by: Jason Wang <jasowang@redhat.com>
10
---
10
---
11
net/tap.c | 10 +++++-----
11
net/tap.c | 2 ++
12
1 file changed, 5 insertions(+), 5 deletions(-)
12
1 file changed, 2 insertions(+)
13
13
14
diff --git a/net/tap.c b/net/tap.c
14
diff --git a/net/tap.c b/net/tap.c
15
index XXXXXXX..XXXXXXX 100644
15
index XXXXXXX..XXXXXXX 100644
16
--- a/net/tap.c
16
--- a/net/tap.c
17
+++ b/net/tap.c
17
+++ b/net/tap.c
18
@@ -XXX,XX +XXX,XX @@ int net_init_bridge(const Netdev *netdev, const char *name,
19
return -1;
20
}
21
22
- fcntl(fd, F_SETFL, O_NONBLOCK);
23
+ qemu_set_nonblock(fd);
24
vnet_hdr = tap_probe_vnet_hdr(fd);
25
s = net_tap_fd_init(peer, "bridge", name, fd, vnet_hdr);
26
27
@@ -XXX,XX +XXX,XX @@ static void net_init_tap_one(const NetdevTapOptions *tap, NetClientState *peer,
28
}
29
return;
30
}
31
- fcntl(vhostfd, F_SETFL, O_NONBLOCK);
32
+ qemu_set_nonblock(vhostfd);
33
}
34
options.opaque = (void *)(uintptr_t)vhostfd;
35
36
@@ -XXX,XX +XXX,XX @@ int net_init_tap(const Netdev *netdev, const char *name,
18
@@ -XXX,XX +XXX,XX @@ int net_init_tap(const Netdev *netdev, const char *name,
19
if (ret < 0) {
20
error_setg_errno(errp, -ret, "%s: Can't use file descriptor %d",
21
name, fd);
22
+ close(fd);
37
return -1;
23
return -1;
38
}
24
}
39
25
40
- fcntl(fd, F_SETFL, O_NONBLOCK);
41
+ qemu_set_nonblock(fd);
42
43
vnet_hdr = tap_probe_vnet_hdr(fd);
44
45
@@ -XXX,XX +XXX,XX @@ int net_init_tap(const Netdev *netdev, const char *name,
26
@@ -XXX,XX +XXX,XX @@ int net_init_tap(const Netdev *netdev, const char *name,
46
goto free_fail;
27
vhostfdname, vnet_hdr, fd, &err);
47
}
28
if (err) {
48
29
error_propagate(errp, err);
49
- fcntl(fd, F_SETFL, O_NONBLOCK);
30
+ close(fd);
50
+ qemu_set_nonblock(fd);
51
52
if (i == 0) {
53
vnet_hdr = tap_probe_vnet_hdr(fd);
54
@@ -XXX,XX +XXX,XX @@ free_fail:
55
return -1;
31
return -1;
56
}
32
}
57
33
} else if (tap->has_fds) {
58
- fcntl(fd, F_SETFL, O_NONBLOCK);
59
+ qemu_set_nonblock(fd);
60
vnet_hdr = tap_probe_vnet_hdr(fd);
61
62
net_init_tap_one(tap, peer, "bridge", name, ifname,
63
--
34
--
64
2.5.0
35
2.7.4
65
36
66
37
diff view generated by jsdifflib
[Qemu-devel] [PULL 3/4] MAINTAINERS: Update the latest email address
[PULL 5/5] net: Use correct default-path macro for downscript
1
From: Zhang Chen <chen.zhang@intel.com>
1
From: Keqian Zhu <zhukeqian1@huawei.com>
2
2
3
Signed-off-by: Zhang Chen <chen.zhang@intel.com>
3
Fixes: 63c4db4c2e6d (net: relocate paths to helpers and scripts)
4
Signed-off-by: Keqian Zhu <zhukeqian1@huawei.com>
4
Signed-off-by: Jason Wang <jasowang@redhat.com>
5
Signed-off-by: Jason Wang <jasowang@redhat.com>
5
---
6
---
6
MAINTAINERS | 2 +-
7
net/tap.c | 3 ++-
7
1 file changed, 1 insertion(+), 1 deletion(-)
8
1 file changed, 2 insertions(+), 1 deletion(-)
8
9
9
diff --git a/MAINTAINERS b/MAINTAINERS
10
diff --git a/net/tap.c b/net/tap.c
10
index XXXXXXX..XXXXXXX 100644
11
index XXXXXXX..XXXXXXX 100644
11
--- a/MAINTAINERS
12
--- a/net/tap.c
12
+++ b/MAINTAINERS
13
+++ b/net/tap.c
13
@@ -XXX,XX +XXX,XX @@ F: include/migration/failover.h
14
@@ -XXX,XX +XXX,XX @@ free_fail:
14
F: docs/COLO-FT.txt
15
script = default_script = get_relocated_path(DEFAULT_NETWORK_SCRIPT);
15
16
}
16
COLO Proxy
17
if (!downscript) {
17
-M: Zhang Chen <zhangckid@gmail.com>
18
- downscript = default_downscript = get_relocated_path(DEFAULT_NETWORK_SCRIPT);
18
+M: Zhang Chen <chen.zhang@intel.com>
19
+ downscript = default_downscript =
19
M: Li Zhijian <lizhijian@cn.fujitsu.com>
20
+ get_relocated_path(DEFAULT_NETWORK_DOWN_SCRIPT);
20
S: Supported
21
}
21
F: docs/colo-proxy.txt
22
23
if (tap->has_ifname) {
22
--
24
--
23
2.5.0
25
2.7.4
24
26
25
27
diff view generated by jsdifflib

Patchew 2.1-devel-b67e4c2

© 2016 - 2025 Red Hat, Inc.